Diversified Trend Trading Approach

USDNOK

20 x R trade :-)



"There and back again" by Bilbo Baggins



Notice how spike was avoided.



USDNOK Result 1 Dec 2018 to 3 May 2018

What is Spacetime?

A fascinating account in layman's terms regarding the cutting edge of physics and the various approaches to quantum gravity seeking to bust wide open the riddle of spacetime.

Awesome mate. There are a few pearls of wisdom growing in this thread that have been demonstrated by the performance metrics produced;
1. That risk can be better managed through a diversified overlay of return streams as opposed to the simple binary decision of placing a stop which forces drawdown into the equation;
2. The concept of noise and signal is a different beast to what is presented visually on a chart by price. Noise is a manifestation of the way price moves over an extended data set whereby a signal can be discerned through serial repetition of a particular price movement . If you take price itself or a pattern of price action as the signal, there is no way to distinguish between periods of serial correlation and periods of simple random overlay.
3. Markets are dynamic and the approach used to tackle them must be adaptive. While the technique can be standardly applied per system, you must adjust for volatility and be prepared to frequently re-balance and re-weight according to your chosen risk-weighted metric (eg. Sortino or MAR etc.).

Notice how all of these features require an extended data set in their approach......... no right or wrong....just different degrees of certainty....and never a simple binary outcome. Sounds like trading to me....but slightly more nuanced than what is presented visually to the brain.

In the words of David Aronson who applied the scientific method to determine the true effectiveness of technical trading signals.....of the 6400 binary buy/sell rules that were rigorously assessed on the S&P500.....none of them worked. You need to dig a bit deeper if you want to trade complexity :-)

Drawing Lines On Charts Is Not A Strategy

USDCHF - A nice strong end to the test. I knew that was coming thanks to R :-)



Performance Results - 1 Dec 2017 to 5 May 2018





Be back in a few hours C :-)

Agree. I truly feel price or any movement of it is the biggest trickster to our eyes. Right after position is entered, I consider Price to be non-existent in any measure. I just let my structured set of exists via Profit/Loss/Margin/Exposure/Drawdown mgmt functions be my guiding principle to manage my portfolio of positions per strategy (what I lack is management across strategies and I am thus far not convinced I need it and that is where you come in :-)
I have cut down all the noise based on sound principles of market micro/macrostructure driven by the world clock (hence Time Action) as I consider Time to be the CRUCIALPOINT in my strategy. i.e Universal Entries/Exits across Time to follow an Actionable Strategy. When it comes to Price and Time, I don't calculate measure or contain or control it. It will always be free-flowing with it's own deceptive personality willing to go whereever it chooses to go. It cannot be mastered or analyzed as it will become invalid the very next unit of measure (tick or millisecond). The only thing I have in my control is my actions that I can take to construct and deconstruct positions in a logical and sensible manner that is within my control. There is a reason why measuring the distance between two trees from a moving train is hard. There is a reason why we all drive in one direction on a highway instead of bi-directional across lanes with no divider.

VEEFX's Comprehensive Solution to Portfolio Diversification via Universal Time Action Strategy

(you probably won't find this in any book)

Time Diversification - via sessions, time of day, time of week, session open, session close, red news, noisy sessions, holidays etc
Price Diversification - via spacing across positions to capture overbought, oversold and equilibrium conditions
Price Behavior Diversification - Price will either Diverge or Converge resulting in Diversion from Mean or Reversion to Mean resulting in either a Breakout or a Bounce. There is no other action involved. If you think there is one, you have been starring at charts for too long!
Duration Diversification - Over 85 Million variables that could affect price at each M1 Candle in any given Month. Approx 450K variables could affect price at the Open of next one minute candle in any given 24 hour cycle. Diversified Duration is accomplished via MTF and over 60+ custom timeframes spanning all timezones around the world
Economic Diversification - Trade all Developed and Emerging Economies. I trade 28 pairs but the principles could be expanded to all economies
Asset Class Diversification - Trade Metals, Commodities and Indexes across market as they are all representative of GLOBAL MONEY FLOWS
Emotion/Sentiment Diversification - Trade ALL red news and they represent the only form to real market sentiment... Capital Flows from Bonds/Treasuries to Risky Assets, Yield curve differential, Fed balance sheet, current accounts, retail/corporate sentiment, embargo and sanctions, import/export imbalance, etc everything going thru the currency markets.
Directional Diversification - Trade Long Only, Short Only and Both Long/Short Neutral direction to spread directional bias
Exposure Diversification - Trade only currencies when the corresponding banks are open. Give enough space between positions curb overcrowding within each pair and within each currency via active position management
Correlation Diversification - Watch for enforced mathematical correlations. Diversify during noise, go head on during fat tails
Outcome Diversification - via Growth and Income. Don't just focus on Income or Growth, I take Income when Market is not moving via 20 times the trading costs and let profit run in active sessions to exploit fat tails as much as possible for Growth
Market Manipulation Diversification - via Law of Large Numbers and Law of Averages. Trade Small and Trade Often to spread the risk of Market Maker working against us. Don't use stoploss or limit orders. It is not your role to feed liquidity to the market. Implement intelligent Risk Management controls and measures to protect the account from unrecoverable drawdowns.
Market Condition Diversification - via trading ranging markets, trending markets, stalled markets and everything in-between based on what Market HAS to do in order for it to survive. Follow the Market Making moves and don't go against it. No amount of analysis can beat market makers.
Trading Style Diversification - via scalping, swing, intra-day and intra-week trading styles. I do it all in one account by letting my portfolio guide me through my universal exits
Broker Technology Diversification - spread accounts across multiple brokers to avoid all eggs in one basket using a single internet connection, power supply etc. Use both desktop and mobile platforms.
Market Closure Diversification - Avoid at all costs. Avoid weekend carryover of positions. We have no control over it so don't do it. Close positions if the instrument stops trading over night (outside of stocks). Most verbal interventions (by design) take place when markets are closed in order to have a bigger impact to price moves within the inner circles of big boys.
Health Diversification - Trading is an extremely unhealthy endeavor. Strike a good balance to ensure you have a live outside of trading. Stare at chart only when need be. I have not looked at charts for over 2 hours unless I have a position on it that I am about to close. My charts are line charts now with foreground same as background so zero visual analysis by design. My EA takes all my commands from my mobile device giving me the freedom to move around and travel.
Association/Bias Diversification - via trade small and trade often. Don't get married to a trade. It is just a blip with no meaning to my overall portfolio. Don't let any one trade lose your sleep over it. All my trades are 1.5% risk on 400:1 for accounts under $1K in size
Markets are Universal and Neutral - This is the only mindset that does not require diversification. This topic is immensely deep so won't get into this here. Thanks to PipEasy, I finally understand the true meaning of markets via his one PM on approaching the market as a data scientist back in 2010.

As CrucialPoint said...

GET IT DONE WITH FOCUS (there is no other way to succeed in trading!)

Love and regards to everyone here at the Factory. After a couple of shout outs, I will exit to continue my journey.

P.S. For those who are addicted to posting, Here's a simple solution to exit the factory (which I think will work... I will know soon once I close my browser)

1. Go to your profile and change your login password to something junk (close your eyes and don't cheat!)
2. Then change your email to something like DSFSDFSDF at DSFHRHFDASD.com
3. Go to your browser history and delete cache and cookies. Repeat these steps on all devices.
4. Close your browser. This should prevent you from logging in and posting. There is a way to even block ff.com url but I won't go that far. After all, you will need some self discipline in this profession.

I will return in within 24 hours and delete this post from my mobile. Adios!

What a cracker of a post to say goodbye with my friend. Given your penchant for leaving no footprints I am going to preserve these fantastic points by quote. I hope you don't mind.

Extract from Veefx

I truly feel price or any movement of it is the biggest trickster to our eyes. Right after position is entered, I consider Price to be non-existent in any measure. I just let my structured set of exists via Profit/Loss/Margin/Exposure/Drawdown mgmt functions be my guiding principle to manage my portfolio of positions per strategy (what I lack is management across strategies and I am thus far not convinced I need it and that is where you come in :-)

I have cut down all the noise based on sound principles of market micro/macrostructure driven by the world clock (hence Time Action) as I consider Time to be the CRUCIALPOINT in my strategy. i.e Universal Entries/Exits across Time to follow an Actionable Strategy. When it comes to Price and Time, I don't calculate measure or contain or control it. It will always be free-flowing with it's own deceptive personality willing to go whereever it chooses to go. It cannot be mastered or analyzed as it will become invalid the very next unit of measure (tick or millisecond). The only thing I have in my control is my actions that I can take to construct and deconstruct positions in a logical and sensible manner that is within my control. There is a reason why measuring the distance between two trees from a moving train is hard. There is a reason why we all drive in one direction on a highway instead of bi-directional across lanes with no divider.

VEEFX's Comprehensive Solution to Portfolio Diversification via Universal Time Action Strategy (you probably won't find this in any book)

Time Diversification - via sessions, time of day, time of week, session open, session close, red news, noisy sessions, holidays etc
Price Diversification - via spacing across positions to capture overbought, oversold and equilibrium conditions
Price Behavior Diversification - Price will either Diverge or Converge resulting in Diversion from Mean or Reversion to Mean resulting in either a Breakout or a Bounce. There is no other action involved. If you think there is one, you have been starring at charts for too long!
Duration Diversification - Over 85 Million variables that could affect price at each M1 Candle in any given Month. Approx 450K variables could affect price at the Open of next one minute candle in any given 24 hour cycle. Diversified Duration is accomplished via MTF and over 60+ custom timeframes spanning all timezones around the world
Economic Diversification - Trade all Developed and Emerging Economies. I trade 28 pairs but the principles could be expanded to all economies
Asset Class Diversification - Trade Metals, Commodities and Indexes across market as they are all representative of GLOBAL MONEY FLOWS
Emotion/Sentiment Diversification - Trade ALL red news and they represent the only form to real market sentiment... Capital Flows from Bonds/Treasuries to Risky Assets, Yield curve differential, Fed balance sheet, current accounts, retail/corporate sentiment, embargo and sanctions, import/export imbalance, etc everything going thru the currency markets.
Directional Diversification - Trade Long Only, Short Only and Both Long/Short Neutral direction to spread directional bias
Exposure Diversification - Trade only currencies when the corresponding banks are open. Give enough space between positions curb overcrowding within each pair and within each currency via active position management
Correlation Diversification - Watch for enforced mathematical correlations. Diversify during noise, go head on during fat tails
Outcome Diversification - via Growth and Income. Don't just focus on Income or Growth, I take Income when Market is not moving via 20 times the trading costs and let profit run in active sessions to exploit fat tails as much as possible for Growth
Market Manipulation Diversification - via Law of Large Numbers and Law of Averages. Trade Small and Trade Often to spread the risk of Market Maker working against us. Don't use stoploss or limit orders. It is not your role to feed liquidity to the market. Implement intelligent Risk Management controls and measures to protect the account from unrecoverable drawdowns.
Market Condition Diversification - via trading ranging markets, trending markets, stalled markets and everything in-between based on what Market HAS to do in order for it to survive. Follow the Market Making moves and don't go against it. No amount of analysis can beat market makers.
Trading Style Diversification - via scalping, swing, intra-day and intra-week trading styles. I do it all in one account by letting my portfolio guide me through my universal exits
Broker Technology Diversification - spread accounts across multiple brokers to avoid all eggs in one basket using a single internet connection, power supply etc. Use both desktop and mobile platforms.
Market Closure Diversification - Avoid at all costs. Avoid weekend carryover of positions. We have no control over it so don't do it. Close positions if the instrument stops trading over night (outside of stocks). Most verbal interventions (by design) take place when markets are closed in order to have a bigger impact to price moves within the inner circles of big boys.
Health Diversification - Trading is an extremely unhealthy endeavor. Strike a good balance to ensure you have a live outside of trading. Stare at chart only when need be. I have not looked at charts for over 2 hours unless I have a position on it that I am about to close. My charts are line charts now with foreground same as background so zero visual analysis by design. My EA takes all my commands from my mobile device giving me the freedom to move around and travel.
Association/Bias Diversification - via trade small and trade often. Don't get married to a trade. It is just a blip with no meaning to my overall portfolio. Don't let any one trade lose your sleep over it. All my trades are 1.5% risk on 400:1 for accounts under $1K in size
Markets are Universal and Neutral - This is the only mindset that does not require diversification. This topic is immensely deep so won't get into this here. Thanks to PipEasy, I finally understand the true meaning of markets via his one PM on approaching the market as a data scientist back in 2010.

As CrucialPoint said...

GET IT DONE WITH FOCUS (there is no other way to succeed in trading!"






Great post mate.



See you in the not too distant future in the FM space mate. You have the right attitude and approach. It has been a pleasure picking apart your mind and has given lots of food for thought. :-) Cheers C

I see you're playing around with Kaufman's KAMA. Do you use the filter he describes in his book "Smarter Trading Improving Performance in Changing Markets"?
I can't figure out how it can be used... The description says that he computes the standard deviation of the KAMA "returns" over 20 periods and a fraction of this is the required minimum opposite move of the KAMA to trigger a reversal. The issue I have is that when the KAMA stops moving in a congestion the SD gets smaller and smaller. The threshols is essentially zero when it is needed big! On the contrary it increases when the KAMA accelerates:


I also read Ramadas is using ATR as a measure of volatility. I personally don't consider ATR to be a valid estimator of the volatility. Judge for yourselves.

Cheers Pip. Just about to hit the sack.

We are just getting our heads stuck into KAMA now. Very early days but it appears to hold some promise over the very short spans we have manually tested on. Hopefully will have an EA up soon to get a bit more rigorous. Whipsaws will need to be dealt with...but starting from scratch to start with simply looking at the crossovers au naturale :-)

I will pick up that book. It has slipped through my net. You are far more advanced in your understanding of the indicator ....but we will get there.

Thanks for the info regarding congestion. Very helpful. I will get deep down and dirty with it and see how it goes.

With regards to volatility measures.....what is your preference? Clearly price movement sequence is an issue with ATR from your diagrams. C

Thank you for your valuable input, Pip.

Please share what would be better alternative to ATR.

My thoughts concerning portfolio construction for Karma:

1. It would need to be uncorrelated static portfolio of let's say 14 markets. Is it diversified enough?
2. If we would allocate max risk of 0.5% per market then whole Karma portfolio heat would be max. 7 %. Is this acceptable heat?

I don't have a definitive preference for the volatility. Thus far the standard deviation of the returns is the best I have. The issue is that to estimate the variance (SD=sqrt(variance)) you need quite a lot of samples (300-500 to have 98-99% confidence) which is unacceptable. With 20 samples the error is HUGE: if you measure x it can be anywhere between x/3 and 3x. I don't have anything better.

Morning mate :-)

Refer to the Houston we have a problem post coming up :-(

C

Thanks for this Pip. I am wondering whether the application of an SDC of a finite bar interval might be a better tool than ATR for volatility adjusting position sizes.

If I apply say a 10 bar SDC (for this thought experiment only) with a defined 1.0 Standard deviation width at each discreet time interval (bar) and then measure the distance in pts between upper and lower bars and then standardise by tick (pts) value for universal use, this might more comprehensively attack volatility. I could then develop a rolling indicator for each progressive bar that would breathe and might be of more use than the ATR.

If I apply this to your visual examples Pip, the SDC channel width should address some of the issues the ATR presents. Your thoughts guys? :-)



Alternatively is it as simple as looking at the high low range of a defined lookback and averaging the results then standardizing in % terms? Do we need to get more complicated than that in addressing volatility? Thoughts guys.

Houston we have a problem

From initial heads exploding....we are now in this position :-)



He who won't be named has developed an EA based on the logic we have applied to the Instant Karma and it just doesn't stack up over a range of market conditions. A false start.....not unusual as we were blinded by our short term samples we tested manually which turns out to be cherry picked and suitable to current market conditions :-(



Sorry guys...can't release the EA but I have attached the fidelity checker between our manual tests and the EA to demonstrate that the EA is working as it should.....but the results of a longer range test on EURUSD will allow you to interpret things and the charts on the fidelity checker will enable you to code up a solution yourselves. Note that all our decisions are made on the close so we can quickly test the logic using control points.

So....it is time now to look at variations on Instant Karma where we attempt to address the whipsaws and the occasional severe loss. .....ideas such as a move to breakeven and let it run..,,,a trailing stop mechanism with an SDC........Thomas's ideas on multiple timeframe decision making on entry .etc. etc. etc.

I still like the logic of KAMA as it can get some exceptional moves but the divergent signals appear to be fleeting by and large with frequent reversion if you let it run unbridled.....so we need to be more refined in addressing them. If the trade doesn't work out quickly....we need to be out. Back to concepts such as trade efficiency like the trusty EDTT. It would be good to develop another system for the portfolio suite so further work needs to be done on KAMA before we call it a day and move on.

In a nutshell....the logic I applied to Instant Karma sux....so time to change the logic without the temptation to optimise.

So thinking caps on again. Them's the breaks.

Cheers

C


PS

Preliminary thoughts.

1. Still apply to every signal but.......

2. Apply a short term SDC to every signal back to the recent high/low (PS this can be coded if a short lookback is all that is required)

3. Use the SDC trailing stop mechanism for trade efficiency

4/ Use the SDC for volatility adjusted position sizing.

Portfolio Results to 30 April 2018 - Fixed Universe

Both EDTT and Reverter went marginally backwards on very little trade volume. In contrast R I think you might be having a little snigger with your dynamic universe for the month.

We need a system to plug Aprils weakness R. We have a long mean reverter and and long/short breakout....now we need a plug in the short reversion space. Have a close look at market conditions in April and see what possibly works on the fixed universe selection.

Take a little chance...it don't mean nothing. Sometimes you win....sometimes you lose :-)



EDTT






Reverter






Blend

Attached Image (click to enlarge)

Cheers

C

Instant Karma - Version 2

We are going to progressively modify the EA from the ground up and test each plank.



Next stage is testing the slow KAMA on a higher timeframe to determine entries and to exit on faster KAMA on lower timeframe. All other variables retained. This is more approaching T's recommendation but we will continue for the moment to keep it trading at all times short and long.

The logic is to see if a tighter exit condition assists in reducing the degree of whipsaw impact.

Stay posted.

C

Use your numbers....but be aware of them as well. Here are some of the traps associated with a bit too much OCD with numbers...... or should that be 123?....... from Mark Rzepctnski from AMPHI :-)



Just remember that MATHS can also mean Mental Abuse To the Human Species. :-)

Dear C

One of the biggest problems we face trying to implement KAMA in Forex markets, is the same nature of those markets.
Kaufman , when designed KAMA, had the stock market in mind.
See below a weekly diagram of NVDA using just our two KAMA combination.
Obviously it works perfectly with a profit of more than 1500% on the last signal only.
But the stock market is a trending market.
Forex is a mean reverting market.
KAMA is a great tool if applied to the correct market.